Rosser Lab

In vivo assays for IntI1 mediated integration and excision

The intI1 gene from the plasmid R388 class I integron has been isolated by PCR and placed under the control of either the IPTG inducible PTRC promoter or the arabinose inducible PBAD promoter in pTRC99a and pBAD33 based plasmids. We have also synthesised totally artificial versions of the intI1 and V. cholerae superintegron intIA1 genes, with single cutting restriction sites engineered every 30-40 bp within the genes. These will allow for simple genetic manipulation of these genes including introduction of point mutations and the production of hybrid proteins.


A bacterial genetic assay has been set up to follow recombination between attC delivered as a single stranded substrate by plasmid conjugation and a plamid-borne attI site resident in the recipient strain. This assay was used to confirm the published result that attC x attI recombination is stimulated when attC is present in single-stranded form, and that this stimulation is strand specific. Recombination was approximately 1000-fold more efficient when the bottom strand of attC was delivered by conjugation than when the top strand was delivered in the same way. When recombinase expression was induced with IPTG, approximately 1 in 10 cells receiving the attC bottom strand integrated it into the attI target plasmid. The rare events that occurred on attC top strands substrates led to integration with the incorrect polarity, as would be expected for strand exchange in attC R'' instead R'.


An assay was also set up for attC x attC recombination in Esherichia coli. A plasmid substrate was constructed containing two attC sites flanking a galK reporter. Recombination between the attC sites leads to loss of the galK reporter, and can be monitored by colony colour on MacConkey galactose plates. Colonies that express galK are red on this medium, whereas those that lack galK are pale. As expected, cells containing the attC x attC reporter plasmid produced red colonies on MacConkey galactose plates. However, introduction of the integrase expression plasmid did not yield the expected white colonies. Examination of plasmid DNA revealed that without induction of integrase expression, approximately 1% of the reporter plasmid had recombined to delete galK, and after induction of integrase this increased to approximately 10% recombination.

However, all cells retained enough copies of the galK+ plasmid to be red on MacConkey galactose indicator plates. We reasoned that this assay could be used to select integrase mutants that recombine attC sites more efficiently than wild-type. These mutants should delete the galK gene by efficient attC x attC recombination and produce pale colonies on MacConkey galactose medium. We made libraries of IntI1 mutants by error prone PCR and introduced them into cells containing the attC galK reporter plasmid. Mutants were obtained that reproducibly produced pale colonies in this assay. Examination of the crystal structure of the closely related V. cholerae superintegron integrase showed that the altered residues in these mutants clustered around important DNA binding regions in IntI. However, these mutants did not lead to an increase in attC x attC recombination, and we believe they are somehow altering the physiology of their host cells so that they produce slightly small pale colonies on MacConkey galactose medium, even in the absence of the attC x attC substrate plasmid.


In vitro assembly using site-specific recombinases.


We have developed a simple system to test the efficacy of sites-specific recombinases for gene assembly in vitro. This system has been used to test a number of different Serine and Tyrosine recombinases for gene assembly. Those genes closest to the promoter are most highly expressed, so different gene orders give different levels of the different enzymes in the metabolic pathway. Cassettes containing regulatory sequences could also be included in the assembly reactions.


Carotenoid biosythetic pathway as a model system for recombinase mediated gene assembly and syntegron shuffling experiments.


We are using the bacterial carotenoid biosynthetic pathway as a model system for our gene assembly and syntegron shuffling experiments. Carotenoids are C40 methyl-branched hydrocarbon compounds built up by successive condensation of C5 (isoprene) units. Beta-carotene is synthesised in carotenogenic bacteria from the C15 compound farnesyl pyrophosphate (FPP) by the addition of a C5 unit to produce the C20 compound geranylgeranyl pyrophosphate (this step is catalysed by CrtE), followed by dimerisation to produce the C40 compound phytoene (catalysed by CrtB), the introduction of four double bonds to produce lycopene (catalysed by CrtI) and finally terminal cyclization to produce beta-carotene (catalysed by CrtY). While phytoene is colourless, lycopene is red and beta-carotene is orange-yellow and their production can easily be detected by bacterial colony colour. The carotenoid precursor FPP is present in non carotenogenic bacteria such as E. coli, so that the introduction of crtE, crtB, crtI and crtY genes is sufficient to produce coloured colonies in E. coli. However, the addition of the idi gene, encoding isopentenyl pyrophosphate isomerase, increases the amount of FPP present in E. coli and increases carotenoid yield. The products of the crtZ and crtX genes process beta-carotene further to zeaxanthin and zeaxanthin diglucoside respectively, the presence of which can be detected by altered colony colour.


So far we have isolated the crtE, crtB, crtI and crtY genes from Erwinia urodevora, and have PCR primers ready to amplify the crtE, crtB and idi genes from Erwinia herbicola. We have placed each of the E. urodevora beta-carotene biosynthetic genes between recombination sites, and are currently using purified recombinases to assemble these genes into synthetic operons containing four or more crt genes in random orders. These synthetic operons will contain multiple crt genes, all in the same orientation, separated by recombination sites, and downstream from a strong regulated promoter. We have just now obtained our first coloured colonies using this recombinase mediated carotenoid pathway assembly. We will determine the gene order in these assemblies by restriction mapping and DNA sequencing, and assay them for carotenoid production in E. coli. Further recombination reactions will allow us to introduce extra genes (such as idi, crtX, and crtZ) into the operon, increasing the product yield or giving different carotenoid products.




We are also using the carotenoid biosynthesis pathway to test our ideas for integron integrase mediated syntegron shuffling in vivo.



  • Dr. Susan Rosser
  • Dr. Sean Colloms
Related publications